Adaptive Consensus ADMM for Distributed Optimization

TL;DR

This paper introduces ACADMM, an adaptive distributed optimization algorithm that automatically tunes parameters for improved performance, achieving a convergence rate of O(1/k) without user intervention.

Contribution

It proposes ACADMM, a novel adaptive consensus ADMM method that automatically adjusts parameters in distributed settings, enhancing efficiency and reliability.

Findings

Achieves O(1/k) convergence rate with adaptive parameters.

Automatically tunes parameters without user input.

Improves distributed optimization performance.

Abstract

The alternating direction method of multipliers (ADMM) is commonly used for distributed model fitting problems, but its performance and reliability depend strongly on user-defined penalty parameters. We study distributed ADMM methods that boost performance by using different fine-tuned algorithm parameters on each worker node. We present a O(1/k) convergence rate for adaptive ADMM methods with node-specific parameters, and propose adaptive consensus ADMM (ACADMM), which automatically tunes parameters without user oversight.